From Web Data to Real Fields: Low-Cost Unsupervised Domain Adaptation for Agricultural Robots

TL;DR

This paper presents a low-cost unsupervised domain adaptation method for agricultural robots, using internet data and a novel attention-based discriminator to improve detection accuracy in new fields.

Contribution

It introduces MAAD, a multi-level attention-based adversarial discriminator, and demonstrates its integration with CenterNet for improved agricultural object detection.

Findings

7.5% increase in object detection accuracy

5.1% improvement in keypoint detection

Effective adaptation from web data to real fields

Abstract

In precision agriculture, vision models often struggle with new, unseen fields where crops and weeds have been influenced by external factors, resulting in compositions and appearances that differ from the learned distribution. This paper aims to adapt to specific fields at low cost using Unsupervised Domain Adaptation (UDA). We explore a novel domain shift from a diverse, large pool of internet-sourced data to a small set of data collected by a robot at specific locations, minimizing the need for extensive on-field data collection. Additionally, we introduce a novel module -- the Multi-level Attention-based Adversarial Discriminator (MAAD) -- which can be integrated at the feature extractor level of any detection model. In this study, we incorporate MAAD with CenterNet to simultaneously detect leaf, stem, and vein instances. Our results show significant performance improvements in the unlabeled target domain compared to baseline models, with a 7.5% increase in object detection accuracy and a 5.1% improvement in keypoint detection.